n8n Expense Receipt Approval: Gmail to Sheets + OpenAI

Turn Gmail expense receipts into approved reimbursement—automatically

This n8n workflow reads new expense receipt emails in Gmail, uses OpenAI to extract structured receipt data, validates each claim against your rules and Google Sheets history, and routes approval requests to the right people—then handles weekly payouts and missing-receipt chasing.

What this workflow does

• Detects incoming receipts: When a new email arrives in a dedicated Gmail inbox, the workflow downloads attachments.
• Loads reference data: It pulls Employees, Policy, prior Claims, and Card transactions from Google Sheets to validate and match claims.
• Extracts receipt fields with OpenAI: Each email is split into individual receipt items (images, PDFs, or forwarded text). OpenAI extracts fields such as dates and amounts, which are normalized for comparison.
• Validates, checks duplicates, and applies policy: The workflow fetches historical exchange rates from the Frankfurter (ECB) API and checks: age/eligibility, currency conversion, company card matches, duplicates, and policy limits to determine reimbursable values and whether manager/finance approval is required.
• Approves or requests review: Within-policy claims are auto-approved; otherwise, the workflow emails the appropriate approver via Gmail using approve/decline action links.
• Logs outcomes and informs employees: It writes one row per receipt line to the Claims tab in Google Sheets with decision details, then emails the employee a line-by-line outcome summary.
• Weekly payout operations: Every Friday, it reads approved unpaid claims and employee banking details from Google Sheets, emails missing-receipt chasers for unmatched company card spend, and sends finance a single Gmail approval to release the payout.

Use cases

• Companies processing frequent employee expense submissions from a shared inbox.
• SaaS operators automating “receipt → extraction → approval → payout” using n8n and Google Sheets.
• Teams needing duplicate detection and card-match verification for policy compliance.

Technical details

• Triggers & logic: n8n nodes including gmail, switch, if, set, code, and no op.
• AI extraction: OpenAI for structured receipt data extraction from images/PDFs/forwarded text.
• Reference data: Google Sheets (Employees, Policy, prior Claims, Card transactions; writes to Claims tab).
• Exchange rates: Frankfurter (ECB) API for historical currency conversion.